Computer Program Development for the Design of IFAS

Wastewater Treatment Processes

Tongchai Sriwiriyarat

Thesis submitted to the Faculty of

Virginia Polytechnic Institute and State University

in partial fulfillment of the requirements for the degree of

Master of Science

in

Environmental Engineering

Clifford W. Randall, Ph.D., Chair

Daniel L. Gallagher, Ph.D.

John C. Little, Ph.D.

April 28, 1999

Blacksburg, Virginia Tech

Keywords: IFAS, activated sludge model, integrated fixed film media, biofilm, mathematical

modeling, nitrification, denitrification

Copyright� 1999, Tongchai Sriwiriyarat

Computer Program Development for the Design of IFAS

Wastewater Treatment Processes

By

Tongchai Sriwiriyarat

ABSTRACT

The Integrated Film Activated Sludge Process (IFAS) was developed to reduce the cost of

additional facilities required to complete year round nitrification in the design of new or retrofit

wastewater treatment plants. The purpose of this project was to develop a computer-based

mechanistic model, called IFAS, which can be used as a tool by scientists and engineers to

optimize their designs and to troubleshoot a full-scale treatment plant. The program also can be

employed to assist researchers conducting their studies of IFAS wastewater treatment processes.

IFAS enables the steady-state simulation of nitrification-denitrification processes as well as

carbonaceous removal in systems utilizing integrated media, but this current version supports

only sponge type media. The IFAS program was developed by incorporating empirical equations

for integrated biofilm carbonaceous uptake and nitrification developed by Sen and Randall

(1995) into the general activated sludge model, developed by the International Association on

Water Quality (IAWQ, previously known as IAWRC), plus the biological phosphorus removal

model of Wentzel et al (1989). The calibration and evaluation of the IFAS model was performed

using existing data from both an IFAS system and a conventional activated sludge bench-scale

plant operated over a wide range of Aerobic Mean Cell Residence Times (Aerobic MCRT’s).

The model developed provides a good fit and a reasonable prediction of the experimental data

for both the IFAS and the conventional pilot-scale systems. The phosphorus removal component

of the model has not yet been calibrated because of insufficient data and the lack of adequately

defined parameters.

ii

ACKNOWLEDGEMENTS

I am sincerely indebted to my major advisor, Dr. Clifford W. Randall, who provided his

invaluable ideas, guidance, suggestions and encouragement throughout my graduate studies,

without whom the IFAS program would not have been developed. Also the expansion of my

professional knowledge and completion of my degree would not have been possible if it were not

by his support, assistance, and inspiration.

I am also grateful to Dr. Daniel L. Gallagher, and Dr. John C. Little for their insights into

my research and their invaluable time to make suggestions and comments. Their sincere

encouragement and guidance inspired me to continue forward throughout my degree program.

Finally, I would like to thank my family for allowing me to pursue my graduate studies

through their financial support, time, and encouragement. Also I would like to thank my close

friends for their friendship and moral support.

iii

TABLE OF CONTENTS

ABSTRACT..................................................................................................................................... i

ACKNOWLEDGEMENTS............................................................................................................ ii

TABLE OF CONTENTS............................................................................................................... iii

LIST OF TABLES.......................................................................................................................... v

LIST OF FIGURES ....................................................................................................................... vi

CHAPTER 1 INTRODUCTION .................................................................................................... 1

CHAPTER 2 LITERATURE REVIEW ......................................................................................... 3

2.1 NUTRIENT EFFECTS ON AQUATIC ECOSYSTEMS ............................................ 3

2.2 BIOLOGICAL NUTRIENT REMOVAL .................................................................... 3

2.2.1 Biological Nitrogen Removal ................................................................................ 4

2.2.2 Biological Phosphorus Removal............................................................................ 4

2.3 BIOFILM PROPERTIES.............................................................................................. 5

2.3.1 Competition in Biofilms ........................................................................................ 5

2.3.2 Simultaneous Nitrification and Denitrification...................................................... 5

2.4 INTEGRATED FIXED FILM BIOLOGICAL PROCESS .......................................... 6

2.5 MEDIA TYPES AND APPLICATIONS ..................................................................... 7

2.6 MATHEMATICAL MODELING................................................................................ 9

2.6.1 Suspended Growth Model...................................................................................... 9

2.6.2 Attached Growth Model ...................................................................................... 10

2.7 NUMERICAL SIMULATION APPROACH............................................................. 12

2.8 REFERENCES ........................................................................................................... 12

CHAPTER 3 COMPUTER PROGRAM DEVELOPMENT FOR THE DESIGN OF IFAS

WASTEWATER TREATMENT PROCESSES .......................................................................... 17

INTRODUCTION ............................................................................................................ 18

MODEL DEVELOPMENT.............................................................................................. 19

iv

PROGRAM DEVELOPMENT AND OPERATION....................................................... 22

MODEL CALIBRATION AND EVALUATION............................................................ 24

RESULTS AND DISCUSSION....................................................................................... 28

CONCLUSIONS............................................................................................................... 38

APPENDIX I. REFERENCES ......................................................................................... 38

APPENDIX II. NOTATION ............................................................................................ 39

v

LIST OF TABLES

Table 1. Stoichiometric parameters for the IFAS model .............................................................. 26

Table 2 Kinetic parameters for the IFAS model at 20oC............................................................. 27

Table 3 Switching function parameters for the IFAS model ........................................................ 28

vi

LIST OF FIGURES

Figure 1 Experimental Data and Model Predictions for SCOD,

IFAS-Sponge System at 12oC…………………………………………………30

Figure 2 Experimental Data and Model Predictions for SCOD,

Control System at 12oC……………………………………………………….31

Figure 3 MLVSS Profiles of IFAS System with UCT/VIP Configuration at 12oC…………..32

Figure 4 MLVSS Profiles of Control System with UCT/VIP Configuration at 12oC………...33

Figure 5 Experimental Data and Model Prediction for NH4+-N, NO3

--N,

IFAS-Sponge System at 12oC at 3.1 aerobic MCRT….………………………34

Figure 6 Experimental Data and Model Prediction for NH4+-N, NO3

--N,

IFAS-Sponge System at 12oC at 1.0 aerobic MCRT…………………………35

Figure 7 Experimental Data and Model Prediction for NH4+-N, NO3

--N,

Control System at 12oC at 3.1 aerobic MCRT………………………………..36

Figure 8 Experimental Data and Model Prediction for NH4+-N, NO3

--N,

Control System at 12oC at 1.0 aerobic MCRT………………………………..37

1

CHAPTER 1

INTRODUCTION

Biological Nutrient Removal (BNR), i.e., the removal of nitrogen and phosphorus from

wastewaters using biological mechanisms, is an advanced wastewater treatment technology that

has been developed to reduce the impacts of nutrients on receiving waters. Eutrophication effects

and algae blooms may be accelerated and stimulated in water bodies as a consequence of nutrient

discharges. Every kilogram of phosphorus assimilated by algae (C106H263O110N16P) typically

generates 111 kilograms of biomass, which is equivalent to 138 kg of COD. If the treatment

plant discharges 5 mg/L of phosphorus, the potential COD formation is 690 mg/L, which

substantially exceeds the 400 mg/L COD concentration of typical untreated USA municipal

sewage. Also a kilogram of nitrogen could potentially produce 16 kilograms of algae, which is

equivalent to 20 kilograms of COD (Randall et al., 1992). The typical treatment plant discharges

about 18 mg/L Total Nitrogen (TN), which can produce 360 mg/L of COD as algae, and almost

equal the COD of raw sewage. The COD thus produced accumulates in the bottom sediments of

lakes, reservoirs, and estuaries when the algae die and settles. This increases the sediment

oxygen demand and accelerates nutrient recycling and eutrophication.

Research and investigations have shown that algal growth in most North American

estuaries is nitrogen limited (Randall et al., 1992 and STAC, 1986), and point source nitrogen

controls are necessary to manage and restore water quality. At many wastewater treatment plants

(WWTP), nitrification and nitrogen removal are seasonal and depends on the growth rate of the

nitrifiers, which is lower in winter than in summer and fall. To obtain complete nitrification all

through the year, the aerobic Mean Cell Residence Time (MCRT) of the biomass must be

increased during winter to exceed the critical MCRT for nitrifiers. However, increasing the

MCRT increases the mixed liquor suspended solids (MLSS) concentration and this may overload

the secondary clarifiers. If so, either the activated sludge basin volume must be increased to

reduce the MLSS concentration going to the clarifiers, or additional clarifier capacity must be

added, unless an innovative approach is used. Recent full-scale and laboratory research has

2

demonstrated that a viable innovative technology is the integration of fixed film media into the

aerobic zone of BNR activated sludge reactors (Sen and Randall, 1995). This technology, labeled

the Integrated Fixed Film Activated Sludge (IFAS) system (Sen, 1995), enhances COD removal

and denitrification, thus enabling year round nitrification and nitrogen removal in activated

sludge systems that are otherwise undersized

.

Mechanistically-based mathematical models have been developed to account for the

major reactions that occur in IFAS systems, to assist engineers or scientists in the development

of optimized design and operation of wastewater treatment systems (Sen, 1995). This is desirable

because IFAS systems are complicated and difficult to analyze because of the large number of

biochemical reactions that occur in the different environmental zones, i.e., anaerobic, anoxic, and

aerobic. Such models reduce the amount of time and the budget needed to explore the feasibility

of the different process options.

The first version of a model called Multiple Cell Wastewater Treatment Plant (MCWTP)

was developed to predict COD removal, nitrification, and denitrification in IFAS systems (Sen,

1995). This model can also be used to compare the performance of IFAS processes with both

conventional BNR and traditional activated sludge processes. However, this version was

developed on a spreadsheet program and for a University of Cape Town/ Virginia Initiative

Project (UCT/VIP) plant configuration. The objectives of this research were to develop a second

mathematical model and a computer program that is more useful and practical. The model was

expected to be more user friendly, interactive, and simpler handling. To be able to characterize

the complex system efficiently, the second version of the IFAS program integrated the empirical

equations for substrate uptake by biofilm on supporting media into the International Association

on Water Quality (IAWQ) general activated sludge model (Model No. 1). The empirical IFAS

equations were modified so that they were compatible with the general model and were able to

reasonably predict the behavior of the fixed film microorganisms. These changes should make

the IFAS program easier to use for practicing design engineers and scientists, and make it

possible to analyze a wide variety of plant configurations.

3

CHAPTER 2

LITERATURE REVIEW

2.1 NUTRIENT EFFECTS ON AQUATIC ECOSYSTEMS

Two primary sources of nutrients, point and non-point, typically contribute nutrients to

receiving waters. An increase of algae and other photosynthetic aquatic life in the water bodies

would be a consequence of increased nutrient inputs. These organisms and plants grow

autotrophically, thus manufacturing organic matter, then eventually die and settle down to the

bottom. Some of the nutrients are then released and recycled back to the water column for new

growth. The algal organic remains can also be consumed by bacteria resulting in the depletion of

dissolved oxygen (DO) in the lower layers of the water column. Also, ammonia-nitrogen formed

by deaminiation of organic compounds can, for instance, be converted to nitrite and nitrate by

nitrifying bacteria with the consumption of DO. The oxygen consumed by this process is

typically referred to as Nitrogenous Oxygen Demand (NOD).

2.2 BIOLOGICAL NUTRIENT REMOVAL

Biological nutrient removal (BNR) can be used to reduce the impact of nutrient discharge

on the oxygen content of the receiving water. It potentially can provide a more economical

alternative of nutrient removal than conventional activated sludge treatment plants using

physicochemical or chemical processes. Excess biological phosphorus removal (EBPR)

eliminates or reduces chemical addition for phosphorus removal. Also, Randall (1991, 1992)

reported that the aeration energy required for BOD stabilization is reduced due to some substrate

utilization that occurs in the anaerobic during BPR. Biological nitrogen removal utilizing the

wastewater BOD for denitrification can result in an aeration energy reduction of as much as

62.5% of the energy required for nitrification (Randall et al., 1992). Both EBPR and biological

nitrogen removal will reduce the amount of sludge that must be processed for disposal, and

4

nitrogen removal will reduce or eliminate the need to add alkalinity for pH adjustment (Randall

et al., 1992).

2.2.1 Biological Nitrogen Removal

Biological Nitrogen Removal consists of two separate sets of reactions to convert

ammonia-nitrogen (NH3N) to gaseous molecular nitrogen (N2). The first set is the

oxidation of the ammonia-nitrogen to nitrite and then to nitrate byNitrosomonasspp. and

Nitrobacterspp., respectively. Typically, the nitrite concentration found in the bulk

solution is less than 0.2 mg/L because the rate of the over all process is limited by

Nitrosomonasspp and nitrite is converted rapidly to nitrate. The second set of reaction

involves the reduction of nitrate to molecular nitrogen. BOD is required for

denitrification, which reduces the amount of DO that must be transferred by aeration if

the wastewater BOD is used for this purpose. Denitrification also produces alkalinity and

can restore up to 50 % of that which was destroyed during the nitrification process, 7.14

mg (as CaCO3) alkalinity consumed per milligram NH3-N oxidized,. 3.57 mg alkalinity

produced per mg of nitrate reduced. Many microorganism species are capable of

denitrifying nitrate in the anoxic zone, but the most commonly observed are

Pseudomonasspecies. Münch et al.(1996) observed that the denitrification rate depends

on the DO concentration and is accurately described by the switching functions presented

in the IAWQ model. They also found that the denitrification rate decreased gradually

with time during aeration periods in sequencing batch reactors (SBRs). It has been

reported that some species may be able to nitrify and denitrify under aerobic conditions.

Robertson et al.(1988) reported thatThiosphaera pantotrophaare able to nitrify NH3-N

and denitrify NO3-N using acetate as a carbon source under aerobic conditions with 30 %

saturation of dissolved oxygen. However, no practical application has ever been

demonstrated.

2.2.2 Biological Phosphorus Removal

5

The Excess Biological Phosphorus Removal (EBPR) method of wastewater

treatment was developed after numerous observations of the accumulation of phosphorus

in excess of typical activated sludge nutritional requirements were reported in the 1960s.

Fuhs et al.(1975) concluded that the principal microorganism group mediating biological

phosphorus removal was theAcinetobactorgenus. These organisms take up short-chain

fatty acids (SCFA) into their cells under anaerobic conditions and store them as poly-β-

hydroxyalkanoates (PHA). The stored PHA is used for energy and growth under

subsequent aerobic conditions, and excess energy is stored in polyphosphate bonds,

resulting in phosphorus removal and storage. The cycle is completed when the energy

stored in the polyphosphate as bonds is used to polymerize the SCFAs as PHAs under

anaerobic conditions. This results in the release of phosphorus to the bulk solution.

Therefore, the anaerobic zone of activated sludge is necessary for biological phosphorus

removal process.

2.3 BIOFILM PROPERTIES

2.3.1 Competition in Biofilm

Studies of biofilm properties have shown that the heterotrophs, nitrifiers, and

denitrifiers coexist in a unit volume of biofilm, increasing with the biofilm depth.

Although the microorganism population increases with biofilm depth, the number of

reactions taking place in the biofilm does not change. From studies, the dissolved oxygen

flux was a key parameter determining the activity of these bacteria. High dissolved

oxygen within biofilm increases the activity of nitrifiers and decreases the amount of

denitrified nitrate (Masuda et al., 1991; Watanabe et al., 1992). Various models have

been developed to characterize the behavior of competition microorganisms in biofilm as

mentioned in the attached growth model section.

2.3.2 Simultaneous Nitrification and Denitrification

6

Masuda et al. (1991) observed the nitrogen loss from a Rotating Biological

Contactor (RBC) plant and hypothesized that there were simultaneous nitrification and

denitrification processes in the biofilm. This occurred when the oxygen transfer rate

decreased enough to create a micro-aerofilic environment. They also demonstrated that

the nitrogen removal efficiency correlated with the temperature, oxygen pressure, influent

organic to ammonia ratio, and hydraulic loading.

Bang et al. (1995) investigated the denitrification process in the aerobic

environment of the biofilm on a RBC. Polyvinyl alcohol (PVA) was used as a carbon

source at 25oC with an influent C/N ratio of 1.2. The net denitrification in the aerobic

zone was about 78%. They also observed that the number of denitrifiers and PVA

decomposing bacteria accumulating in the surface layer were higher than those in the

middle and bottom layers, by up to 1 to 2 orders of magnitude. The nitrification

efficiency was directly proportional to the oxygen flux. The denitrification efficiency, on

the other hand, was indirectly proportional to the oxygen flux to the biofilm layers. They

verified this experiment with a computer simulation. Adequate agreements between both

results were obtained (Watanabe et al., 1992).

2.4 INTEGRATED FIXED FILM BIOLOGICAL PROCESS

The integrated fixed film activated sludge (IFAS) process was introduced to enhance

carbonaceous and nutrogen removal. This technology combines the main advantages of the

attached growth system and the suspended growth system, and minimizes their disadvantages.

The attached growth system provides a high volumetric density of microorganisms accumulating

by natural attachment as biofilm on the media. Shock load and toxic load problems can be

minimized by this process due to the large amount of biomass on the integrated media. This

increases the cell retention time without increasing the mixed liquor concentration in the system

and overloading the secondary clarifier. These are potential problems with suspended growth

biological system. However, suspended growth treatment systems are more flexible than fixed

film systems for meeting higher effluent quality standards particularly nutrient removal. IFAS

process, however, is a hybrid system which provides both stability and flexibility.

7

Researchers have reported that temperature has a primary impact on the kinetics of

various biochemical conversions in the biological nutrient removal processes that result in

substantially lower growth rates of the microorganisms. Large wastewater treatment plants

located in the temperate zones of North American were observed and found that they fall in the

range of 10-12oC (Sen, 1995). McClintock et al. (1991,1993) also found that EBPR could not be

completed at 5 days MCRT and 10oC. More adverse effects on the performance of EBPR and

nitrogen removal due to low temperature can be due to low MCRT. Mamais and Jenkins (1992)

reported that EBPR functioned efficiently at 2.9 days aerobic MCRT for the temperature range

investigated (13.5-20oC). Temperature affected EBPR when the system MCRT was operated

below 2.9 days. The growth rate of nitrifiers is dramatically lower at temperatures of 10-12oC as

in winter compared to summer when temperatures exceed 20oC (Ekama and Marais, 1984). A

primary thrust of the Chesapeake Bay Program is to incorporate BNR into the existing

conventional activated sludge plants as economically as possible. To accomplish year round

nitrification, it is necessary to increase the quantity of biomass maintained in smaller the aerobic

zone. This can be done by increasing the Mixed Liquor Suspended Solid (MLSS) concentration ,

but this will result in higher sludge loadings to the clarifiers, and may exceed their capacity.

Alternatively, the aeration basin volume can be increased. Therefore, larger aeration basins or

clarifier tanks are required. Due to the constraints economic resources and space, the integrated

fixed film activated sludge was an option which was evaluated and shown that can minimize the

size needs of the aeration tank or clarifier. This is because a large amount of microorganisms can

be maintained on the support media without substantially increasing loading to the clarifier.

2.5 MEDIA TYPES AND APPLICATIONS

Several types of media can be utilized in integrated fixed film activated sludge treatment

systems. Ringlace, a rope type media, was installed in the wastewater treatment plant of the city

of Geiselbullach, Germany to upgrade the wastewater treatment plant to achieve a year round

nitrification. This purpose of this upgrading was to maintain nitrification throughout the year and

improve sludge settling and it has successfully accomplished. (Lessel, 1991; Lessel, 1993). In the

United States, Ringlace IFAS was installed in the treatment plant of the city of Annapolis,

8

Maryland and operated as both a Modified Ludzack Ettinger (MLE) and a Step Feed BNR

process. (Sen and et al., 1993; Sen and et al., 1994a; Sen, 1995; Randall and et al., 1996; and

Randall, 1997). The data showed that the integrated Ringlace enhanced nitrogen removal. In both

plant configurations, nitrification in the IFAS section was more than three times as much as in

the control section without media. Also, the amount of aerobic denitrification was enhanced in

the IFAS system when operation was changed from MLE to Step Feed BNR.

Sponge media was tested in the Geiselbullach, Germany by Lessel (1991), but was not

selected as a biofilm media because of its anaerobic sludge production. Randall (1997)

categorized applications of floating sponge media into three types. For the first type, the media

zone was placed after primary sedimentation and operated without recycles before the activated

sludge zone. This was used at Moundsville, West Virginia, USA by Golla et al.(1993) with

Captor as the sponge media. There was 70-75% nitrogen removal in the Captor zone over a

three-year monitoring period. The average ammonia nitrogen effluent was 5.4 mg/L with a

standard deviation of 3.6 mg/L, and was below the effluent ammonia nitrogen standard. For the

second type, Linpor-N , another floating sponge media designed to enhance nitrification, was

installed as a last treatment step without subsequent clarification at Wyk, Germany (Morper and

Wildmoser, 1989; Schonberger,1989). At least 50% nitrogen removal was achieved even at a

temperature as low as 10oC. For the last concept, the floating media was added directly to the

aeration tank of the activated sludge process to enhance nitrification in winter. This system was

successfully implemented by Sen and Randall (1994) at the Cox Creek Wastewater treatment

plant, Maryland, USA. Also a bench-scale plant with sponge IFAS obtained almost complete

nitrification while only 50% was obtained in a control system with no sponges. In Japan, Kondo

(1992) conducted an experiment of porous media application in a domestic wastewater treatment

plant. Effluent BOD and nitrogen concentrations of less than 20 and 15 mg/L were obtained over

water temperature range of 13-31oC could be achieved as long as the optimum pore size and pad

dimensions were chosen. This plant was operated with only 1.5 hours of aeration and a

subsequent 0.5 hours of agitation. Tsuno and et al., (1992) investigated the applicability of

polyurethane foam cubes in the sewage treatment plant at Kyoto City, Japan. Polyurethance

foam cubes of 8 cm3 were added into an aerobic reactor consisting of 6 cells that operated

9

without recycle. The nitrification rate could reached 0.33 mg N/hr-cube at a temperature of 15oC.

Jensen (1995) found that freely floating media significantly increased nitrification at a

temperature of 12oC under operating conditions of low aerobic MCRT (3.4 to 1.7 days).

Ringlace did not perform nitrification as well as Captor media. However, denitrification in the

aerobic reactor was higher with the Ringlace media (Mitta, 1994). Liu (1996) confirmed that

nitrification could be maintained in the Captor system at 12oC at aerobic MCRTs as low as 1.7

days. There was at least 15 mg/L difference in effluent ammonia between the IFAS and the

control system at the selected aerobic MCRT. It was shown that the IFAS system could maintain

nitrification at an aerobic MCRT and temperature at which the control system would fail.

2.6 MATHEMATICAL MODELING

The complexity of activated sludge systems has expanded from system designed for

carbonaceous compounds removal only to include nitrification, denitrification and excess

biological phosphorus removal. Many biochemical reactions and compounds are involved in

these processes. The computer model plays an important role in the simplification of the

complexity of wastewater treatment processes for both full-scale and bench-scale systems. In

full-scale operation, the model can be used to optimize the processes or troubleshoot the

treatment systems. It can be applied at the bench-scale in research studies for testing results,

optimizing processes, or selecting processes. The computer models mainly are divided into two

categories according to the type of biological growth, i.e., suspended growth or attached growth.

2.6.1 Suspended Growth Model

A task group was formed by IAWQ (previously known as IAWPRC) in 1982 to

facilitate the application of models for designing and operating activated sludge systems

and many models have since been developed to describe various complex activated

sludge systems. Grady (1986) initially presented a single-sludge system model, which

10

consisted of seven processes including carbon oxidation, nitrification, and denitrification,

which was the outline for a model that was agreed on by the task group. The SSSP was

developed to describe a single sludge system with the matrix approach proposed by

IAWQ (Bidstrup, 1988). A preliminary model was evaluated by Dold and Marais(1986)

and this led to the development of Activated Sludge Model No. 1 (ASM1) development.

Modifications were made and adopted as a final version of ASM1 (Henze and et al.,

1987). Subsequently, many other models have been developed building upon the ASM1

model. ASM1 does, however, not include the phenomenon of EBPR. Extension of ASM1

to include EBPR was proposed and developed by IAWQ and called ASM2. ASM2

includes both the excess biological phosphorus removal processes and chemical

processes for phosphorus precipitation (Henze et al., 1994). Barker and Dold (1997)

developed their general model, which includes 36 processes, among them, the EBPR

component. The anoxic growth of poly-P microorganisms also has been incorporated in

the model. Anoxic and anaerobic hydrolysis of slowly enmeshed biodegradable COD

processes are incorporated into this model as well as the fermentation of readily

biodegradable organic materials. The general model is based on the original ASM1 and

the Wentzel Model (1989). Henze (1999) proposed ASM No. 2d that includes

phosphorus removal by of poly-P microorganisms during denitrification. Gujer (1999)

presented an ASM No. 3, which is a modification of ASM No. 1. A substrate storage

process was introduced into this version as a new process and it is assumed that there is

no COD flow from nitrifier decay. The modeling of heterotroph and autotroph decay

processes were clearly separated in ASM3.

2.6.2 Attached Growth Model

Two factors have been raised when dealing with the modeling of fixed growth

systems. First, the effects of diffusive resistance on substrate removal need to be

incorporated into the model to accurately predict the performance of the system. Second,

the heterogeneous nature and poor characterization of these conditions within the biofilm

result in the difficulty of the biofilm modeling difficulties. Early mathematical models

focused on substrate utilization in the biofilm. Substrate utilization was described as a

11

process of molecular diffusion and simultaneous biochemical reactions (Williamson and

McCarty, 1976). Numerous steady state models have been developed, which vary from a

single microorganism species to multiple-species and from a single substrate to multiple-

substrates. Subsequently, a steady state biofilm model of a single substrate coupled with

substrate flux into the biofilm has been developed by assuming neither net growth nor

decay over time. A minimum substrate concentration in the bulk liquid must be

maintained to achieve a steady state condition. The substrate flux and thickness are

assumed to be zero when the substrate concentration is below this criteria. The Monod

equations are used to describe biochemical reactions and Fick’s law is used to describe

diffusion equations, which combined together are the basic mechanisms in biofilms.

Idealized biofilms have been described as having uniform cell density and uniform

thickness. All nutrients required for growth are assumed to be in excess concentrations

(Rittmann and McCarty, 1980a). The model was evaluated and the results showed that

the model successfully predicted the substrate utilization rate and biofilm

thickness.(Rittmann and McCarty, 1980b). Mass transport by diffusion is assumed to be

one dimensional because the biofilms are typically very thin. Due to the possibility of

extensive computational efforts, steady state mechanistic models with multiple-species

with competition of substrate and space were developed. A multi-species model for COD

removal and nitrification was proposed by Wanner and Gujer (1984), but this model was

not verified with experimental data. Wanner and Gujer (1984) showed that given the

competition for oxygen, the heterotrophs may be washed out at low substrate

concentrations. In contrast, at high substrate concentrations, they may overgrow the slow-

growing autotrophic organisms. The heterotrophs have a tendency to reside at the surface

and autotrophs may accumulate more in the deeper layers. The biomass flux was used to

simulate biomass growth. However, this model is not a true steady state model.

Subsequently, Rittmann and Manem (1992) presented a steady state model that described

multi-species behavior with multiple active species, inert biomass, substrate utilization,

diffusion within the biomass, external mass transport, and the detachment phenomenon.

This model was verified experimentally and its predictions agreed with the observed data

in terms of space competition. Higher acetate concentrations resulted in NH4+N flux

decrease because autotrophs were forced deeper into the biofilm by heterotrophs. Kissel

12

et al.(1984) proposed a dynamic model, allowing the volume of biomass to increase with

time from V to V+∆V at time t to t+∆t when biomass increases due to positive net

growth.

2.7 NUMERICAL SIMULATION APPROACH

The equations applied in the activated sludge model form a system of non-linear

equations. Many numerical approaches are available from the literature to solve these equations.

However, the approach that can be used to solve most conditions is preferred. The global

convergence method was used since it is reported in the literature that it guarantees the

convergence of these equations with any starting point (Press et al., 1992). Although more

computational needs are required, the Modified Newton-Raphson method, combines the rapid

local convergence of Newton’s method and the global convergent strategy, it is a powerful

method because it guarantees some degree of progression toward the solution at each iteration.

2.8 REFERENCES

Bang, D.Y., Watanabe, Y., and Noike, T. (1995). “An experimental study on aerobic

denitrification with polyvinyl alcohol as a carbon source in biofilms”. Wat. Sci. Tech.,

32(8), 235-242.

Barker, P.S., and Dold, P.L. (1997). “General model for biological nutrient removal activated

sludge systems: model presentation”. Water Environment Research., 69(5), 969-984.

Bidstrup, S.M. and Grady Jr., C. P. L. (1988). “SSSP --simulation of single-sludge processes”.

J. Wat Pol. Con. Fed. 60(3), 351-350.

Dold, P.L., and Marais, G.v.R (1986). “Evaluation of the general activated sludge model

proposed by the IAWPRC task group. Wat. Sci. Tech., 18, 63-89.

Ekama, G.A. and Marais, G.v.R. (1984). “Chapter 5: Nitrification”. In: Theory, Design and

Operational of Nutrient Removal Activated Sludge Processes”. Water Research

Commission, Pretoria, SA.

Fuhs, G.W., and Chen M. (1975). “Microbiological basis of phosphate removal in the activated

sludge process for the treatment of wastewater”. Microbiol. Ecology, 2, 119-138.

13

Grady, C.P.Jr., Gujer, W., Henze, M., Marais G.v.R., and Matsuo, T. (1986). “A model for

single-sludge wastewater treatment systems”. Wat. Sci. Tech., 18, 47-61.

Golla, P.S., Reddy, M.P., Simms, M.K., and Laken, T.J. (1993). “Three years of full scale

Captor process operation at Moundsville, WWTP”. Proc. 2nd International Conference

on Biofilm Reactors, IAWQ, Paris, Frace, 239-245.

Gujer W., Henze, M., Mino, T. and Van Loosdrecht, M.C.M (1999). “Activated sludge model

No.3”. Wat. Sci. Tech., 39(1), 183-193.

Henze, M., Grady, C.P.L., Gujer, W., Marais, G.v.R., and Matsuo, T. (1987). “ Activated sludge

model No.1”. Sci. Tech. Report No.1, Int. Assoc. Water Pollut. Res. Control, London,

U.K.

Henze, M., Gujer, W., Mino, T., Matsuo,T.,Wentzel, M.C., and Marais, G.v.R. (1994).

“Activated sludge model No.2.” Sci. Tech. Report No.3, Int. Assoc. Water Qual.,

London, U.K.

Henze, M., Gujer, W., Mino, T., Matsuo, T., Wentzel, M.C., Marais, G.v.R., and Van

Loosdrecht, M.C.M.(1999). “Activated sludge model No. 2d, ASM2D”. Wat. Sci. Tech.,

39(1), 165-182.

Jensen, K.R. (1995). “Effects of integrated fixed film activated sludge on nitrogen removal in

biological nutrient removal systems”. MS Thesis, Virginia Tech, Blacksburg, VA 24061.

Kissel, J.C., McCarty, P.L., and Street, R.L.(1984). “Numerical simulations of mixed-culture

biofilm”. J. Environ.Engr. (ASCE) 110, 393-411.

Kondo, M. Hozo, S., and Inamori, Y. (1992). “Simultaneous removal of BOD and nitrogen with

anoxic/oxic porous biomass support system”. Wat. Sci. Tech., 26(9-11), 2003-2006.

Lessel, T.H. (1991). “First practical experience with submerged rope type (Ringlace) biofilm

Reactors for upgrading and nitrification”. Wat. Sci. Tech, 23, 825-834.

Lessel, T.H. (1993). “Upgrading and nitrification by submerged biofilm reactors – Experiences

from a large scale plant”. Proc.: 2nd International Specialized Conf. On Biofilm Reactors.

IAWQ, Sep 29-Oct 1, Paris, 231-238.

Liu, H. (1996). “Utilization of captor sponges to maintain nitrification and denitrification in

BNR activated sludge at low aerobic MCRTs”. MS Thesis, Virginia Tech, Blacksburg,

VA 24061.

Mamais, D. and Jenkins, D.. (1992). “The effects of MCRT and temperature on enhanced

14

biological phosphorus removal”. Wat. Sci. Tech., 26(5/6), 955-965.

Masuda, S., Watanabe, Y., and Ishiguro, M. (1991). “Biofilm properties and simultaneous

nitrification and denitrification in aerobic rotating biological contactors”. Wat. Sci. Tech.,

23, 1355-1363.

McClintock, S. A., Randall,C.W., Pattarkine,V.M. (1993). “Effect of temperature and Mean

Cell Residence Time on biological nutrient removal processes”. Water Environment

Research, 65(2), 110-118.

McClintock, S.A., Randall, C.W., and Pattarkine, V.M.(1991). “Effects of temperature and mean

cell residence time on enhanced biological phosphorus removal”. Proc. Specialty Conf.

on Environ. Eng. J. Environ.Eng., Reno, NV.,319-324.

Mitta, P.R. (1994). “Utilization of fixed film media in BNR activated sludge systems”. MS

Thesis, Virginia Tech, Blacksburg, VA 24061.

Morper, M., and Wildmoser, A. (1989). “Improvement of existing wastewater treatment plants

efficiencies without enlargement of tankage by application of the Linpor-Process – Case

studies”. Proc. Upgrading of Wastewater treatment plants. IAWPRC, International

Specialized Conference, Munich, Germany, 3-7.

Mÿnch E.V., Lant P., and Keller J. (1996). “Simultaneous nitrification and denitrification in

bench-scale sequencing batch reactors”. Water Research, 30(2), 277-284.

Press, W.H, Teukolsky, S.A., Vetterling, W.T., and Flannery, B.P.(1992). “Numerical recipes in

C: The art of scientific computing”. 2nd , Cambridge University Press, 383-393.

Randall, C.W., Brannan, K.L, McClintock,S.A., and Pattarkine, V.M. (1991). “The case for

anaerobic reduction of oxygen requirements in biological phosphorus removal system”.

Water Envir. Res., 64(6), 824-833.

Randall, C.W., Barnard, J.L., Stensel, H.D.(1992). “Design and retrofit of wastewater treatment

plants for biological nutrient removal”. Water Quality Management Library. Vol.5.

Technical Publishing Company, Inc., Allentown, PA.

Randall, C.W., and Sen, D.(1996). “Full scale evaluation of an integrated fixed film activated

sludge (IFAS) process for enhanced nitrogen removal”. Wat. Sci. Tech., 33(12), 155-162.

Randall, C.W. (1997). “Activated Sludge: Lastest developments and a look into the future”.

15

CIWEM Conference: Activated Sludge into the 21st Century”. The University of

Manchester, England. 17-19 September 1997. 30 pages. CIWEM Services Ltd. 15 John

Street, London, WCIN 2EB, England.

Rittmann, B.E., and McCarty, P.L. (1980a). “Model of steady-state-biofilm kinetics”.

Biotechnol. Bioeng., 22, 2343-2357.

Rittmann, B.E., and McCarty, P.L. (1980b). “Evaluation of steady-state-biofilm kinetics”.

Biotechnol, Bioeng., 22, 2359-2373.

Rittmann, B.E., and Manem, J.A. (1992). “Development and experimental evaluation of a steady

state, multispecies biofilm model”. Biotechnol Bioeng, 39, 914-922.

Robertson, L.A., et al. (1988). “Simultaneous nitrification and denitrification in aerobic

chemostat cultures ofThiosphaera pantotropha”. Appl. and Env. Microbiol. 54, 2812.

Sen, D., Farren, G.D., Copithorn, R.R., and Randall, C.W., (1993). “Full scale evaluation of

nitrification and denitrification on fixed film media (Ringlace) for design of single sludge

system”. Proc: Water Environment Federal, 66th Annual Conference, Volume 3-Liquid

Treatment Symposium, 137-148, Anaheim, CA.

Sen, D., and Randall, C.W.(1994a). “Performance of fixed film media integrated in the activated

sludge reactors to enhance nitrogen removal”. Wat.Sci.Tech.,30(11), 13-24.

Sen, D., Randall, C.W., Jensen, K., and et al. (1994b). “Design parameters for integrated fixed

film activated sludge (IFAS) processes to enhance biological nitrogen removal”. Proc:

Water Environment Federal, 67th Annual Conference, Volume 1-Part1 Biological

Treatment Systems and Part 2 Biological Nutrient Removal, Chicago, IL.

Sen, D.(1995). “COD Removal, Nitrification and Denitrification kinetics and mathematical

Modeling of an Integrated Fixed Film Activated Sludge (IFAS) System”. Ph.D.

Dissertation, Virginia Tech, Blacksburg, VA 24061.

Schonberger, R. (1989). “Conversion of existing primary clarifiers according to the EASC

process for biological phosphorus removal”. Proc. Upgrading of Wastewater treatment

plants. IAWPRC, International Specialized Conference, Munich, Germany, 3-7.

Tsuno, H., Somiya, I., Matsumoto, N., and Sasai, S.(1992). “Attached growth reactor for BOD

removal and nitrification with polyurethane foam medium”. Wat. Sci. Tech.,26(9/11).

2035-2038.

Watanabe, Y., Masuda, S., and Ishiguro, M. (1992) “Simultaneous nitrification and

16

denitrification in micro-aerobic biofilms”. Wat. Sci. Tech, 26(3/4), 511-522.

Wanner O. and Gujer W. (1985). “Competition in biofilms”. Wat. Sci. Tech., 17, 27-44.

Wentzel, M.C., Dold, P.L., Ekama, G.A., and Marais, G.v.R. (1989) “Enhanced polyphosphate

organism cultures in activated sludge systems. III, Kinetic Model.” Water SA, 15, 89

Williamson, K., McCarty, P.L.(1976a). “A model of substrate utilization by bacterial films”. J.

Water Pollut. Control Fed., 48(1), 9-24.

17

Computer Program Development for the Design of IFAS

Wastewater Treatment Processes

By T. Sriwiriyarat1 and C.W. Randall2

Abstract : The Integrated Fixed Film Activated Sludge Model (IFAS) is a computer-based

mechanistic model package that allows steady-state simulation of IFAS wastewater treatment

processes. IFAS wastewater treatment systems are activated sludge biological nutrient removal

processes that have been enhanced by the addition of biofilm support media into the aerobic zone

of the system. This process is used to obtain year round nitrification in activated sludge systems

that otherwise could not support it. The IFAS model was developed based on the IAWQ general

model for activated sludge and empirical equations for COD uptake and nitrification on biofilm

developed at Virginia Tech. The current version of the IFAS Model supports only sponge media;

however, the model could be modified for other media if the appropriate equations and

parameters are known. Data obtained from an IFAS sponge pilot scale plant treating a weak

municipal wastewater supplemented by sodium acetate and phosphates were used to calibrate the

model. The calibrated model was then verified with data obtained from experiments using other

influent wastewater concentrations and different operating conditions. The model generated

ammonia and soluble COD profiles that were insignificantly difference statistically from the

experimental data. Enhanced biological phosphorus removal also has been incorporated into the

model, but has not yet been verified. Verification studies are being planned and will be

performed in the future. The IFAS Model satisfactorily predicts carbonaceous removal and

nitrification, and has the potential to be a useful tool for scientists and engineers seeking to

design and optimize either IFAS or conventional BNR activated sludge systems. It also could be

used to troubleshoot full-scale treatment plants.

1Graduate Student, Civil and Environmental Engineering Department,

418 NEB, Virginia Tech, Blacksburg, VA 24061.2The C.P. Lunsford Professor, Civil and Environmental Engineering Department,

418 NEB, Virginia Tech, Blacksburg, VA 24061.

18

INTRODUCTION

The growth rate of nitrifiers in the winter, i.e., at low temperatures, is substantially lower

than in the summer or fall, i.e., at higher temperatures. To obtain year round complete

nitrification with conventional activated sludge systems in temperate zones, the amount of

biomass maintained in the aerobic zones must be increased in the winter relative to what is

required in the summer and fall. This is usually accomplished by increasing the solids retention

time (SRT) of the mixed liquor suspended solids (MLSS) in the activated sludge reactors. This

increases the MLSS concentration, which often overloads the secondary clarifiers. Then it is

necessary to either increase the clarifier capacity so that more solids can be processed, or to

increase the total reactor volume so that the MLSS concentration to the clarifiers is reduced.

However, an alternative approached has been developed in recent years, the Integrated Fixed

Film Activated Sludge (IFAS) process (Lessel, 1993, Sen and Randall, 1994, Randall and Sen,

1996). This technology integrates biofilm support media into activated sludge aerobic reactors to

increase the biomass concentration and enhance nitrification. Experience has shown that it also

enhances denitrification in the aerobic zone. The coexistence of suspended growth and attached

growth enhances the amount of biomass in the reactor without contributing a substantially to the

amount of sludge flowing forward to the secondary clarifier because the additional biomass

remains attached to the support media, which is retained in the aerobic reactor.

Extensive pilot-scale studies of IFAS processes have been conducted by Virginia Tech

researchers to evaluate different types of media, elucidate the fundamentals of the IFAS

mechanisms, and determine the impacts of IFAS on biological nutrients removal systems (Mitta,

1994; Jensen, 1995, Sen, 1995, Liu, 1996). These studies evaluated both rope-like Ringlace

media and Captorsponge media. Additionally, the Ringlace media was evaluated and utilized

for the upgrade of the 37,850 m3/d Annapolis, Maryland Water Reclamation Facility, and the

Captor media was evaluated at the 56,775 m3/d Cox Creek Facility near Baltimore, Maryland

(Sen, et al., 1993; Randall, 1996). All of these investigations showed that IFAS could be used to

enhance both nitrification and denitrification in activated sludge systems without significantly

increasing the solids loading to the secondary clarifiers.

19

Modeling tools that assist engineers or scientist to optimize their designs or to operate

wastewater treatment plants (WWTP) are needed to predict the fate of components in the

activated sludge system, as well as treated effluent. The models allow exploration through a

simulation of alternative wastewater treatment plant configurations, inputs, and operational

strategies. Once the model is calibrated to a particular wastewater, the model can be applied to

screen the potential process designs and to eliminate the inefficient and uneconomic designs. The

objective of this project was to develop, a user-friendly computer-based model that could be used

to evaluate IFAS systems and compare them with conventional BNR and fully aerobic actiavted

sludge systems. The goals were to develop a computer-based mechanistic and empirical IFAS

model that would forecast the simultaneous COD removal, nitrification, and denitrification

processes occurring in a properly functioning IFAS system, and enable comparison with non-

IFAS activated sludge system.

MODEL DEVELOPMENT

The model developed was designed to provide the same benefits as the mechanistic

general model for biological nutrient removal activated sludge systems, initially developed at the

University of Cape Town, adapted by the International Association on Water Quality (IAWQ) as

Activated Sludge Model No. 1 (ASM1), and subsequently modified by Barker and Dold (1997).

The rate expressions used were a combination of those employed by Barker and Dold (1997) for

their adaptation of the general model, and empirical IFAS equations developed by Sen (1995)

and co-workers. The ASM1 model is a mechanistic model for carbonaceous organic removal,

nitrification, and denitrification (Henze, 1987), which was subsequently modified to ASM2 by

incorporating the excess biological phosphorus removal model developed by Wentzel, et al.

(1989) at the University of Cape Town. The details of the general model will not be discussed in

this paper, but the reader is referred to the literature cited above.

The amount of substrate uptake by biofilm on integrated media was established by the

empirical equations developed by Sen and co-workers. The mass balance equations were then

established with the appropriate rate expressions. The IFAS program solves these mass balance

20

equations simultaneously as a system of nonlinear equations using the Modified Newton

techniques. Various assumptions were made to simplify the complex activated sludge system

model, which are listed as follows:

- No diffusion of nitrate from the bulk solution into the biofilm. Therefore, the

concentration of nitrate in the mixed liquor may be slightly higher than the true

concentration in the aerobic zone because nitrate is not denitrified by the biofilm.

- No excess biological phosphorus removal will be accomplished by the biomass

growth on the supporting media. Only normal phosphorus uptake for biofilm growth

is incorporated into the equations.

- Sufficient alkalinity is assumed to be provided to the system in the simulation of the

model. Therefore, the alkalinity component is not included in the mechanistic

equations.

- No nitrite accumulation occurs in the system. All nitrite nitrogen is assumed to be

rapidly converted to nitrate byNitrobactorspp., except when DO is limiting.

- No biomass in the incoming influent wastewater. Therefore, no influent biomass

effects on the performance of the system.

- pH is constant and near neutrality to ensure optimum biological activity.

Nitrification Rates for Biofilm on Sponge Cuboids

The maximum ammonium-nitrogen (NH4+-N) uptake rate of the sponge biofilm

(qm,Nitr,sp) was determined at low soluble biodegradable COD (SCODbio) levels to

minimize nitrogen uptake by the heterotrophs (Sen ,1995; Sen et al., 1995). Ammonium-

N levels, greater than 10 mg N/L, and dissolved oxygen (DO) above 7 mg/L were

provided to achieve zero order nitrification rates with respect to ammonium-N and DO. It

was established by Sen (1995) that the relationship between SCODbio and NH4+-N was

such that the NH4+-N uptake rate was inhibited if the concentration of SCODbio was

above 10 mg/L, as shown in Equation 1. A possible explanation is space competition

pressure as defined by the multi-species biofilm theory (Rittmann and Manem, 1992, and

Wanner and Gujer, 1985). The higher specific growth rate heterotrophs utilize SCODbio

21

and grow at the surface layer of the biofilm. The slow-growing nitrifiers are forced

deeper into the biofilm layer and experience greater mass transport resistance,

particularly a inadequate supply of DO.

10bioSCODSNKSNKN4NHA

sp,Nitr,mq−+

−= (1)

where ANH4-N = 4.4 mg NH4+-N per sponge cuboid per day at 12oC, and

KSN = Half saturation coefficient = 9.4 mg/L SCODbio at 12oC

When the SCODbio is less than 10 mg/L, the nitrification rate is controlled by the NH4+-N

concentration in the mixed liquor. It is represented by Equation 2.

N4NH055.1sp,Nitr,mq −+= (2)

COD Uptake Rate for Biofilm on Sponge Cuboids

Batch tests were carried out with sponge media and mixed liquor to determine a

maximum COD uptake rate for sponge biofilm (qm, COD, sp). The determination by Sen

(1995) from the experimental data showed that the rate was related to the SCODbio in the

mixed liquor as follows:

CbioSCODSMK

bioSCODbioSCODAsp,COD,mq +

+= (3)

When C = 0 mg SCOD/sponge/day, which was the best fit,

ASCODbio= 51.5 mg SCOD per sponge per day at 12oC

KSM = Half saturation coefficient = 19.3 mg/L SCODbio at 12oC

22

The overall rate equation for nitrification was adapted from Barker and Dold, 1997, and the

number relates back to their numbering system for rate equations. It combines nitrification in the

mixed liquor and on the biofilm, and can be expressed as follows:

SpongeDOBF,AUT,OK

DO

N4NHBF,NHK

N4NHsp,Nitr,mqAY

AZDOSS,AUT,OK

DO

N4NHSS,NHK

N4NHA16r

���

�

���

�

+���

�

���

�

−+−

+���

�

���

�

+���

�

���

�

−+−µ=

(4)

The overall aerobic growth rate of heterotrophs with acetate as a carbon source and ammonium

nitrogen as a nitrogen source is:

SpongeDOBF,HET,OK

DO

BSASBF,SHKBSAS

sp,COD,mqHY

HZP4POSS,GROK

P4PO

N4NHSS,NAK

N4NH

DOSS,HET,OK

DO

BSASSS,SHKBSAS

H1r

���

�

���

�

+���

�

���

�

+

+���

�

���

�

−+−

���

�

���

�

−+−

���

�

���

�

+���

�

���

�

+µ=

(5)

The above growth rate is also applied for process rates using SBSC as the substrate and NO3- as

the electron acceptor (Barker and Dold, 1997).

At steady state, the biofilm detachment rate must be equal to the biofilm growth rate.

Therefore, excess microorganisms from biofilm growth are detached to the mixed liquor. The

degree of detachment depends on the turbulence in the mixed liquor. First order decay rates are

used for both the suspended growth and the attached biomass.

PROGRAM DEVELOPMENT AND OPERATION

IFAS was written for IBM-PC compatible microcomputers in Borland C++ Builder suite.

Borland C++ Builder package was chosen because it provides the visual productivity of Delphi

while it permits work in a C++ environment. The system was designed for utilization with a 32

23

bit Windows based program. IFAS program was written and compiled on Windows 98 and is

compatible with Windows 95/98 or Windows NT. Many functions were added to the program to

make it user-friendly and interactive to users. IFAS allows the user to enter the specific operating

and design data, in addition to the wastewater characteristics. The system MCRT is used instead

of the aerobic MCRT because the system MCRT is a more general design parameter than aerobic

MCRT. The user is able to select either an IFAS or a Conventional Activated Sludge system.

The CAS system of this program is a general activated sludge plant configuration without

media integrated in the aerobic zone, and may be either a BNR process or a conventional fully

aerobic process. Default kinetic and stoichiometric parameters are provided as well as switching

function coefficients to allow simulations for investigative or educational purposes. All

parameters must be modified for the specific wastewater characteristics and the mode of

operation selected. The functions related to the IFAS process will be disabled if a conventional

activated sludge system is selected. The process configuration can be defined by recycle flows,

the mixed liquor DO concentration, and the number of reactors. The program allows simulation

of tanks in series up to 12 activated sludge reactors in the treatment train. The volume of each

reactor must be entered when the process configuration is defined. IFAS then utilizes all

information to set up mass balances equations and simulates the fate of all defined components

in the wastewater treatment plant. The sludge wastage rate is computed automatically based on

the system MCRT. All particulate biomass and inert organic matters entering the secondary

clarifiers are assumed to be returned to the process in recycle streams. It is assumed that there is

no accumulation of solids with time and that no reactions occur in the clarifier. It also is assumed

that there are no particles discharged in the effluent. After the wastage rate is known, the steady

state equations are solved simultaneously by the Modified Newton technique using the globally

convergent method.

The global technique converges a solution from almost any initial guess or starting point.

The system MCRT is employed to compute the initial conditions based on a single complete

mixed reactor without sponge media installed. However, the IFAS program will request that the

MCRT be reentered with a value different than the MCRT for wastage flow calculation, if the

solution does not converge. The results can be presented in several ways after the solution has

24

converged, i.e., as line or bar charts indicating the substrate profile, as a complete printable

detailed report, and as a table of all substrate concentrations in each reactor. All results can be

exported to different formats such as Excel and HTML to make them easier to manipulate. In

other words, the data can be manipulated by different programs even if the computer does not

have the IFAS program installed.

MODEL CALIBRATION AND EVALUATION

A large number of kinetic and stoichiometric parameters taken from the literature and

experimental data were used to calibrate the IFAS model. Pilot plant experiments treating a weak

municipal wastewater supplemented by sodium acetate and phosphates were conducted using

pilot-scale systems located on the Blacksburg campus of Virginia Tech (Sen, 1995). Wastewater

was pumped from an adjacent sewer, and then stored for 24 hours inside the small building

containing the pilot systems. The 24 hour storage period was used to bring the wastewater to

temperature equilibrium, accomplish removal of any DO and nitrates contained in the sewage,

and to allow some fermentation of the organic compounds. Because the wastewater was weak in

biodegradable fermentable organics, acetate was added until it was approximately 66 percent of

the influent COD. This percentage of acetate was used to describe the influent wastewater

characteristics in the model predictions.

The two pilot plant systems simulated the UCT/VIP process configuration. Sponges were

added to the aerobic zone of one of the systems for IFAS evaluation, and the other was used as a

BNR control. Both treatment trains consisted of two anaerobic tanks, two anoxic tanks, and three

aerobic tanks in series. Experiments were carried out at different aerobic MCRTs and operating

conditions. Data from these experiments were utilized to establish the yield coefficient (YH), the

decay rate (bH), the maximum substrate utilization rates by the MLSS and the biofilm (qmH), and

the half saturation constant (KSH) for the heterotrophs. The maximum growth rate (µmax) and YH

determined from these experiments were converted to COD units to consistency with COD mass

balance determinations and in conformance with the IAWQ models.

25

The influent wastewater in these experiments was fully characterized into COD, TKN,

and TP fractions, i.e., biodegradable, nonbiodegradable, ammonia, nitrate, etc., for input into the

program. Other kinetic, stoichiometric and switching function parameters for the three types of

bacteria in BNR systems were taken from the general model (Barker and Dold, 1997). Some

parameters were adjusted to obtain the best fit of the model to the experimental data. A single set

of kinetic and stoichiometric parameters produced accurate predictions for both systems and five

different aerobic MCRTs per system. The calibrated values of the stoichiometric, kinetic, and

switching function parameters are listed in Table 1 to 3. The parameters associated with excess

phosphorus removal are not presented in this paper.

26

Table 1. Stoichiometric parameters for the IFAS model

Symbol Characterization Value Units

Non-polyP Heterotrophs

Yhaer Aerobic yield 0.582 mg cell COD yield (mg COD utilized)-1

Yhanx Anoxic yield 0.440 mg cell COD yield (mg COD utilized)-1

Yhana Anaerobic yield 0.440 mg cell COD yield (mg COD utilized)-1

Yac Fermentation Sbsa yield 0.700 mg Sbsa COD (mg Sbsc COD)-1

Eanox Anoxic hydrolysis efficiency factor 0.700 mg Sbsc COD (mg Senm COD)-1

Eana Anaerobic hydrolysis efficiency

factor

0.500 mg Sbsc COD (mg Senm COD)-1

Fnzh Nitrogen content of active mass 0.085 mg N (mg COD active organisms)-1

Fnzeh Nitrogen content of endogenous mass 0.085 mg N (mg COD endogenous residue)-1

Fpzh Phosphorus content of active mass 0.021 mg P (mg COD active organisms)-1

Fpzeh Phosphorus content of endogenous

mass

0.021 mg P (mg COD endogenous residue)-1

Feph Fraction of active mass remaining as

endogenous residue 0.080 mg COD endogenous mass (mg COD

active mass)-1

Fcvh COD:VSS ratio 1.420 mg COD (mg VSS)-1

Autotrophs

Ya Yield 0.142 mg cell COD yield (mg N utilized)-1

Fnza Nitrogen content of active mass 0.085 mg N (mg COD active organisms)-1

Fnzea Nitrogen content of endogenous mass 0.085 mg N (mg COD endogenous residue)-1

Fpza Phosphorus content of active mass 0.021 mg P (mg COD active organisms)-1

Fpzea Phosphorus content of endogenous

mass

0.021 mg P (mg COD endogenous residue)-1

Fepa Fraction of active mass remaining as

endogenous residue 0.080 mg COD endogenous mass (mg COD

active mass)-1

Fcva COD:VSS ratio 1.420 mg COD (mg VSS)-1

27

Table 2 Kinetic parameters for the IFAS model at 20oC

Symbol Characterization Value Units θθθθT

Non-polyP Heterotrophs

µH Maximum specific growth rate 6.423 d-1 1.030

KSH,SS Half saturation coefficient for suspended

solids growth

60.805 mg COD L-1 1.030

KSH,BF Half saturation coefficient for biofilm

growth

60.085 mg COD L-1 1.030

BH Decay rate 0.064 d-1 1.055

ηS,ANX Anoxic solubilization factor 1.000

ηS,ANA Anaerobic solubilization factor 0.500

ηGRO Anoxic growth factor 0.496

KH Maximum specific hydrolysis rate 4.450 d-1 1.050

KX Half saturation coefficient for hydrolysis 0.150 mg COD (mg COD)-1 1.000

KC Maximum specific fermentation growth

rate

4.000 d-1 1.029

KS,ANA Fermentation half saturation growth rate 5.000 mg COD L-1 1.000

KR Ammonification rate 0.080 L (mg COD.d)-1 1.029

ASCODbio Maximum utilization rate on sponge 65.239 mg SCODbio(sponge.d)-1 1.030

KSM Half saturation coefficient for COD

utilization

24.423 mg SCOD L-1 1.030

Autotrophs

µA Maximum specific growth rate 0.545 d-1 1.030

KNH,SS Half saturation coefficient for mixed

liquor growth

1.495 mg N L-1 1.060

KNH,BF Half saturation coefficient for biofilm

growth

1.495 mg N L-1 1.060

BA Decay rate 0.035 d-1 1.060

ANH4-N Maximum nitrification rate on sponge 5.574 mg NH4N (sponge.d)-1 1.030

KSN Half saturation coefficient on sponge 14.982 mg SCOD L-1 1.060

28

Table 3 Switching function parameters for the IFAS model

Symbol Characterization Value Units

Mixed Liquor Section

KO,HET,SS DO limit for heterotrophs in mixed liquor 0.700 mg O2 L-1

KO,AUT,SS DO limit for autotrophs in mixed liquor 0.500 mg O2 L-1

KNA,SS Ammonia limit in mixed liquor 0.005 mg N L-1

KNO,SS Nitrate limit in mixed liquor 0.100 mg N L-1

KGRO,SS Phosphate limit in mixed liquor 0.005 mg P L-1

Biofilm Section

KO,HET,BF DO limit for heterotrophs in biofilm 2.000 mg O2 L-1

KO,AUT,BF DO limit for autotrophs in biofilm 3.000 mg O2 L-1

KNA,BF Ammonia limit in biofilm 2.000 mg N L-1

RESULTS AND DISCUSSION

Comparison of the experimentally determined substrate profiles with predicted values

from IFAS showed that the IFAS model predictions produced accurate profiles for all systems

and different operating conditions. Most estimated soluble COD and ammonium-N profiles fell

within the range of values of the experimental data as shown in Figure 1, 2, 5, 6, 7, and 8. That

is, the IFAS predictions were statistically insignificantly different from the experimental data.

However, as expected, the predicted nitrate profiles were higher in the aerobic zones of the IFAS

sponge system because it was assumed that no nitrate in bulk solution diffused to the biofilm. In

fact, it is know that nitrates do diffuse into the biofilm and this results in substantial

denitrification.

Figure 1 presents the soluble COD profiles in the IFAS system at 3.1 and 1.0 day aerobic

MCRTs as were determined by experiment, and compares them with predictions by Sen’s

spreadsheet IFAS model, and the modified IFAS model of this research. A much better

prediction of the soluble COD profile was obtained with the modified IFAS model compared to

29

Sen’s spreadsheet model by incorporating a hydrolysis term for the particulate COD. The SCOD

profiles in the control system for the same operating conditions are shown in Figure 2.

Figures 3 and 4 show the MLVSS profiles for both systems. It was noted that the

predictions by both models for the last anoxic and first two aerobic reactors (Figure 4) did not

quite agree with the experimental data. It is likely that there were analytical errors during the

experiments that resulted in the observed differences.

In Figures 5, 6, 7, and 8, the predicted ammonium-nitrogen from IFAS program

compared well with the experimental data. This was an improvement over the spreadsheet

version, which included the biodegradable organic nitrogen in the profiles. In the IFAS

predictions, all slowly biodegradable substrate and biodegradable organic nitrogen were

accounted for by including them in the SCODbio and NH4+-N profiles through hydrolysis

processes. However, both versions can predict satisfactorily over the aerobic zones for both the

SCOD and ammonium profiles because the fractions of particulate COD and organic nitrogen

were quite low in the aerobic reactors. Further investigation of substrate profiles in the biofilm is

needed to model the amount of nitrate denitrified in the biofilm.

30

Reactor No.

Influent 1 2 3 4 5 6 7

So

lubl

eC

OD

(Exp

,Exc

el,I

FA

S)

(mg/

L)

0

100

200

300

400

500

600

Experimental DataExcel = SCOD+ P-CODbioIFAS Model Prediction

Reactor No.

Influent 1 2 3 4 5 6 7

Sol

uble

CO

D(E

xp,E

xcel

,IF

AS

)(m

g/L

)

0

100

200

300

400

500

600

Experimental DataExcel = SCOD + P-CODbioIFAS Model Prediction

UCT/VIP IFAS System at 3.1 day Aer MCRT

UCT/VIP IFAS System at 1.0 day Aer MCRT

Anaerobic Anoxic Aerobic

Anaerobic Anoxic Aerobic

TCOD

TCOD

Figure 1. Experimental Data and Model Predictions for SCOD,IFAS-Sponge System at 12oC

31

Reactor No.

Influent 1 2 3 4 5 6 7

Sol

uble

CO

D(E

xp,E

xce

l,IF

AS

)(m

g/L)

0

100

200

300

400

500

600

Experimental DataExcel = SCOD+P-CODbio

IFAS Model Prediction

Reactor No.

Influent 1 2 3 4 5 6 7

Sol

uble

CO

D(E

xp,E

xce

l,IF

AS

)(m

g/L

)

0

100

200

300

400

500

600

Experimental DataExcel = SCOD+P-CODbio

IFAS Model Prediction

UCT/VIP Control System at 1.0 day Aer MCRT

UCT/VIP Control System at 3.1 day Aer MCRT

Anaerobic Anoxic Aerobic

Anaerobic Anoxic Aerobic

TCOD

TCOD

Figure 2. Experimental Data and Model Predictions for SCOD,Control System at 12oC

32

Reactor No.

Influent 1 2 3 4 5 6 7

ML

VS

S(m

g/L

)

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Experimental MLVSS at 3.1 day Aer MCRTIFAS MLVSS Prediction at 3.1 day Aer MCRTExperimental MLVSS at 1.0 day Aer MCRTIFAS MLVSS Prediction at 1.0 day Aer MCRT

Anaerobic Anoxic Aerobic

3.1 days Aer MCRT

1.0 day Aer MCRT

Figure 3. MLVSS Profiles of IFAS System with UCT/VIP Configuration at 12oC

33

Reactor No.

Influent 1 2 3 4 5 6 7

ML

VS

S(m

g/L

)

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Experimental MLVSS at 3.1 days Aer MCRTIFAS MLVSS Prediction at 3.1 days Aer MCRTExperimental MLVSS at 1.0 day at Aer MCRTIFAS MLVSS Prediction at 1.0 day Aer MCRT

Anaerobic Anoxic Aerobic

3.1 days Aer MCRT

1.0 day Aer MCRT

Figure 4. MLVSS Profiles of Control System with UCT/VIP Configuration at 12oC

34

Anaerobic Anoxic Aerobic

Reactor No.

Influent 1 2 3 4 5 6 7

NH

4+-N

,Uno

xN(m

gN

/L)

0

10

20

30

40

50

60

70

80

NO

3- -N(m

gN

/L)

0

2

4

6

8

10

12

Experimental NH4+-N

ExcelUnoxN = NH4+-N + NBS + NBP

IFAS NH4+-N Prediction

IFAS NO3--N Prediction

Experimental NO3--N

TKN

UCT/VIP IFAS System at 3.1 day Aer MCRT

Figure 5. Experimental Data and Model Prediction for NH4+-N, NO3

--N, IFAS-Sponge System at 12oC

35

Influent 1 2 3 4 5 6 7

0

10

20

30

40

50

60

70

80

NO

3- -N(m

gN

/L)

0

2

4

6

8

10

Experimental NH4+-N

ExcelUnoxN = NH4+-N + NBS + NBP

IFAS NH4+-N Prediction

Experimental NO3--N

IFAS NO3--N Prediction

NH

4+-N

,Uno

xN(m

gN

/L)

Anaerobic Anoxic Aerobic

TKN

UCT/VIP IFAS System at 1.0 day Aer MCRT

Figure 6. Experimental Data and Model Prediction for NH4+-N and NO3

--N, IFAS-Sponge System at 12oC

Reactor No.

36

Reactor No.

Influent 1 2 3 4 5 6 7

0

10

20

30

40

50

60

70

80

0

2

4

6

8

10

Experimental NH4+-N

ExcelUnoxN = NH4+-N + NBS + NBP

IFAS NH4+-N Prediction

Experimental NO3--N

IFAS NO3--N Prediction

NH

4+-N

,Uno

xN(m

gN

/L)

NO

3- -N(m

gN

/L)

Anaerobic Anoxic Aerobic

TKN

UCT/VIP Control System at 3.1 day Aer MCRT

Figure 7. Experimental Data and Model Predictions for NH4+-N and NO3

--N, Control System at 12oC

37

Influent 1 2 3 4 5 6 740

45

50

55

60

65

70

0.0

0.2

0.4

0.6

0.8

1.0

Experimental NH4+-N

ExcelUnoxN = NH4+N + NBS + NBP

IFAS NH4+-N Prediction

Experimental NO3--N

IFAS NO3--N Prediction

NH

4+-N

,Uno

xN(m

gN

/L)

NO

3- -N(m

gN

/L)

Anaerobic Anoxic Aerobic

TKN

Figure 8. Experimental Data and Model Predictions for NH4+-N and NO3

--N, Control System at 12oC

UCT/VIP Control System at 1.0 Aer MCRT

Reactor No.

38

CONCLUSION

An IFAS computer model package that predicts carbonaceous removal, nitrification, and

denitrification at steady state conditions in activated sludge systems with either integrate sponge

media (IFAS) in the aerobic zone or without integrated media was developed. The model is

based on the general model of Barker and Dold (1997) integrated with empirical equations

developed by Randall, Sen and co-workers at Virginia Tech from IFAS pilot-scale and full-scale

studies. The model was calibrated using kinetic and stoichiometric coefficients from the

literature and the experimental pilot plant studies. The model predictions agreed very well with

the experimental data from the pilot studies. The soluble COD and NH4+-N concentrations

obtained using the IFAS program closely agreed with the observed results, which was a

substantial improvement over the spreadsheet IFAS version developed by Sen (1995). The

nitrate profiles could be further improved by incorporating the nitrate flux into the biofilm into

the program, but further experimental work is required before this will be possible

The developed model has the potential to be a useful tool for engineers or scientists who

desire to design or operate IFAS systems to obtain year round nitrification. The IFAS program

also can be used to predict the location and amount of sponge media needed to optimize the

design and operation of IFAS systems.

APPENDIX I. REFERENCES

Barker, P.S., and Dold, P.L. (1997). “General model for biological nutrient removal activated

sludge systems: model presentation”. Water Environment Research., 69(5), 969-984.

Henze, M., Grady, C.P.L., Gujer, W., Marais, G.v.R., and Matsuo, T. (1987). “ Activated sludge

model No.1”. Sci. Tech. Report No.1, Int. Assoc. Water Pollut. Res. Control, London,

U.K.

Jensen, K.R. (1995). “Effects of integrated fixed film activated sludge on nitrogen removal in

biological nutrient removal systems”. MS Thesis, Virginia Tech, Blacksburg, VA 24061.

Liu, H. (1996). “Utilization of captor sponges to maintain nitrification and denitrification in

39

BNR activated sludge at low aerobic MCRTs”. MS Thesis, Virginia Tech, Blacksburg,

VA 24061

Mitta, P.R. (1994). “Utilization of fixed film media in BNR activated sludge systems”. MS

Thesis, Virginia Tech, Blacksburg, VA 24061.

Randall, C.W., and Sen, D. (1996). “Full-scale evaluation of an integrated fixed-film activated

sludge (IFAS) process for enhanced nitrogen removal.” Wat. Sci. Tech. 33(12), 155-162.

Sen, D., Farren, G.D., Copithorn, R.R., and Randall, C.W. (1993). “Full scale evaluation of

nitrification and denitrification on fixed film media (Ringlace) for design of single sludge

system”. Proc: Water Environment Federal, 66th Annual Conference, Volume 3-Liquid

Treatment Symposium, 137-148, Anaheim, CA.

Sen, D. (1995). “ COD removal, nitrification, and denitrification kinetics and mathematical

modeling of an integrated fixed film activated sludge (IFAS) system.” Ph.D. Thesis,

Dept. of Civil and Environmental Engineering, Virginia Tech, Blacksburg, Virginia.

Sen, D., Randall, C.W., and Liu, H. (1995). “Effect of suspended solids MCRT on COD and

nitrogen removal kinetics of biofilm support media integrated in activated sludge

systems”. Proc: Water Environment Federal, 68th Annual Conference & Exposition,

Session No. 34, Miami, FL.

Wentzel, M.C., Dold, P.L., Ekama, G.A., and Marais, G.v.R. (1989) “Enhanced polyphosphate

organism cultures in activated sludge systems. III, Kinetic Model.” Water SA, 15, 89

APPENDIX II. NOTATION

Aer MCRT = Aerobic MCRT (day)

DO = Dissolved oxygen (mg O2/L)

NBP = Particulate biodegradable organic nitrogen (mg N/L)

NBS = Soluble biodegradable organic nitrogen (mg N/L)

NH4+-N = Ammonium nitrogen (mg N/L)

NO3--N = Nitrate nitrogen (mg N/L)

MLVSS = Mixed Liquor Volatile Suspended Solids (mgCOD/L)

VSS = Volatile Suspended Solids (mg MLVSS/L).

P-CODbio = Particulate biodegradable COD (mg COD/L)

40

PO4-P = Soluble orthophosphate (mg P/L)

SCFA = Short Chain Fatty Acid (mg COD/L)

SBSA = Soluble readily biodegradable SCFA COD (mg COD/L)

SBSC = Soluble readily biodegradable complex COD, non-SCFA (mg COD/L)

SCOD = Total Soluble COD = SBSA + SBSC + SUS(mg COD/L)

SCODbio = Soluble biodegradable COD = SBSA + SBSC (mg COD/L)

Sponge = Number of sponge cuboid (Sponge/L)

SUS = Soluble unbiodegradable COD (mg COD/L)

TCOD = Total COD (mg COD/L)

TKN = Total Kjeldahl nitrogen (mg N/L)

UCT/VIP = University of Cape Town/Virginia Initiative Project

UnoxN = Unoxidized Nitrogen = NH3-N/NH4+-N + NBP + NBS (mg N/L)

ZA = Autotrophic organism mass (mg COD/L)

ZH = Non-polyP heterotrophic organism mass (mg COD /L)